Académique Documents
Professionnel Documents
Culture Documents
Vyjeyanthi S. Periyakoil, MD,1,2 Eric Neri, MS,1 and Helena Kraemer, PhD1
Abstract
Background: To provide preference-sensitive care, we propose that clinicians might routinely inquire about
their patients’ bucket-lists and discuss the impact (if any) of their medical treatments on their life goals.
Methods: This cross-sectional, mixed methods online study explores the concept of the bucket list and seeks to
identify common bucket list themes. Data were collected in 2015–2016 through an online survey, which was
Downloaded by 191.248.94.42 from online.liebertpub.com at 02/21/18. For personal use only.
completed by a total of 3056 participants across the United States. Forty participants who had a bucket list were
identified randomly and used as the development cohort: their responses were analyzed qualitatively using
grounded theory methods to identify the six key bucket list themes. The responses of the remaining 3016
participants were used for the validation study. The codes identified from the development cohort were vali-
dated by analyses of responses from 50 randomly drawn subjects from the validation cohort. All the 3016
validation cohort transcripts were coded for presence or absence of each of the six bucket list themes.
Results: Around 91.2% participants had a bucket list. Age and spirituality influence the patient’s bucket-list.
Participants who reported that faith/religion/spirituality was important to them were most likely (95%) to have a
bucket list compared with those who reported it to be unimportant (68.2%), v2 = 37.67. Six primary themes
identified were the desire to travel (78.5%), desire to accomplish a personal goal (78.3%), desire to achieve
specific life milestones (51%), desire to spend quality time with friends and family (16.7%), desire to achieve
financial stability (24.3%), and desire to do a daring activity (15%).
Conclusions: The bucket list is a simple framework that can be used to engage patients about their healthcare
decision making. Knowing a patient’s bucket list can aid clinicians in relating each treatment option to its
potential impact (if any) on the patient’s life and life goals to promote informed decision making.
Keywords: advance care planning; bucket list; cross cultural; goals of care; multi ethnic; personalized medicine
1
Division of Primary Care and Population Health, Center of Population Health Sciences, Stanford University School of Medicine,
Stanford, California.
2
VA Palo Alto Health Care System, Palo Alto, California.
Accepted December 20, 2017.
1
2 PERIYAKOIL ET AL.
bucket list, and whether they were able to list specific items was assessed for 100% of the development cohort transcripts;
on their bucket list. The data were analyzed to examine for agreement was found to be greater than 90%, which was deemed
common bucket list themes. to be comparable with previous work. After 35 transcripts had
been coded, no new codes emerged from the next five transcripts
Materials and Methods data indicating that saturation had been reached. At the end
Survey design of this process, the primary themes were identified, and the
development cohort was discarded and not used in subsequent
In April 2015, we deployed a survey to better understand analyses. Actual words and phrases that participants used were
diverse Americans’ attitudes toward end-of-life issues. It was used to name the themes as feasible (Table 2 for exemplars).
developed by one of the authors (V.S.P.) based on prior The responses of the validation cohort were first reviewed
studies.11,12 The survey was beta-tested with a small group of for the presence or absence of bucket list items. The bucket
patients and families and their suggestions were used to refine list codes identified from the development cohort were vali-
and finalize the questions. The project was reviewed and dated by analyses of responses from 50 randomly drawn
approved by the Institutional Review Board at Stanford Uni- subjects from the validation cohort. For those participants in
versity School of Medicine. the validation cohort who submitted a bucket list, their re-
sponses were coded independently by the two coders for the
Data collection
presence or absence of the individual primary themes iden-
The survey was housed online on a secure Stanford server and tified by qualitative analyses of the development cohort.
a link to it was posted on a dedicated online portal. Information Coders compared codes for inter-rater reliability. All dis-
about the project was disseminated through the Stanford web crepancies in codes were reviewed by both coders with one of
portals13–15 and through articles in mainstream media15–21 the authors (H.C.K.) and discussed until consensus was
publicizing the project. The survey was administered once and reached. The data were imported into SAS SAS 9.4, SAS
no personal health identifiers were collected in an effort to Institute (Cary, NC) for quantitative analyses.
promote participant confidentiality and honest responses
without concerns about individual scrutiny. The secure online Quantitative analyses
system was programmed to prevent ballot box stuffing. All Participants with a bucket list were compared to those who
questions in the survey were set at ‘‘force response’’; partic- did not have one using a chi-square test for categorical variables
ipants who did not respond to all the questions would be un- (e.g., race and gender) and the Mann-Whitney-Wilcoxon test for
able to submit the survey. Sociodemographic characteristics continuous variables (e.g., age) and ordinal variables (e.g., level
including, age, gender, ethnicity/race, and self reported im- of education and relative importance ascribed to faith/religion/
portance ascribed to faith/religion/spirituality spiritual affili- spirituality). For participants who had a bucket list, recursive
ation are shown in Table 1. The participants’ state of residence partitioning using Quality Receiver Operating Curve (QROC)
was determined using their zip codes. The data presented here analysis24,25 helped explore differences between groups of
were collected from July 2015 through December 2016. The participants and identify subgroups with common patterns.
investigators had no direct contact with the participants.
Results
Question prompt used for data collection
Based on their self-reported place of residence, we had par-
‘‘The term ‘‘bucket list’’ refers to a list of things that one ticipants from all the fifty states in the United States. Of the 3016
has not done before but wants to do before dying. Do you validation cohort participants, a vast majority (91.2% i.e., 2752
have a bucket list? If yes, please list the items on your bucket out of 3016) reported having a bucket list. In comparing the
list if any in the order of importance.’’ Participants could group with a bucket list with those who did not have one,
enter upto five bucket list items in text boxes provided. analyses identified that age (Mann-Whitney-Wilcoxen v2 of
COMMON ITEMS ON A BUCKET LIST 3
Table 1. Participants Demographics Comparing the Group with a Bucket List with Those without One
Does not have Comparison of those
a bucket list Has a bucket list without and with a bucket list
Mean Mean MWW MWW
Description N (SD) N (SD) Chi-Square p-value
Age in years 264 56.5 (14.3) 2752 50.0 (16.0) 39.01 <0.0001
MWW MWW
English is the language spoken at home Chi-Square p-value
39.01, p < 0.0001), and gender (v2 of 40.19, p < 0.0001) and prevalent in women <33 years of age (69.3%) and
marital status (v2 of 13.8, p = 0.0032) were influential factors. least in unmarried persons ‡59 years of age (22.3%).
Participants who reported that faith/religion/spirituality was (4) Desire to spend quality time with friends and family was
important to them were most likely (95%) to have a bucket the fourth prevalent (16.7%) theme. Participants who
list compared with those who reported it to be unimportant were ‡63 years of age were most likely to list this desire.
(68.2%), v2 = 37.67 (Fig. 1). Women younger than 70 years of (5) Desire to achieve financial stability was the fifth prevalent
age, for whom faith/religion/spirituality was at least somewhat theme with an overall prevalence of 16.1% across all
important, were most likely to have a bucket list at 94.9%. ethnic groups and significantly higher in African Ameri-
Participants ‡61 years of age, who were not married and for cans (24.3%).
whom faith/religion/spirituality was not important were least (6) Desire to do a daring activity was the last theme with a
likely (31.8%) to have a bucket list. prevalence of 15%. Younger participants (<26 years of
Six primary themes were identified by analyzes of the age) were exponentially more likely (28.9%) to report this
respondents’ bucket list: desire on their bucket list compared to older participants
(‡61 years of age) at only 7.6%.
(1) Desire to travel, within the nation or internationally, was
the most prevalent item on the bucket list (78.5%).
Discussion
People most likely to list travel were college-educated
women (84.3%), followed by men <65 years of age for To the best of our knowledge, our study is the first to sys-
whom faith was unimportant (80.6%). Unmarried men tematically investigate the concept of the bucket-list, an idea of
‡65 years were least likely to list travel (52.3%) on growing interest to the general public. This concept, if harnessed
their bucket list. thoughtfully, has great potential in engaging patients about their
(2) Desire to accomplish a personal goal 78.3% of the health behaviors and health-related decision making by using a
participants identified this desire on their bucket list, framework that they can understand easily. Through our study we
and there were no discriminating subgroups. have shown that adults and older adults from diverse backgrounds
(3) Desire to achieve specific life milestones was the are able to articulate their bucket list. Our study shows that those
third prevalent (51%) theme. This category was most who ascribe more importance to spirituality are more likely to
Table 2. Common Bucket List Themes and Verbatim Exemplars
Bucket list themes Exemplars
Desire to travel ‘‘Go to Japan’’
‘‘Backpack through Europe’’
‘‘Go to Hawaii’’
‘‘Visit Australia’’
‘‘Visit Cuba’’
‘‘Cycle in italy and France’’
‘‘See Niagara Falls’’
‘‘Visit presidential libraries’’
‘‘See Anne Frank House & Corrie ten Boom Museum’’
‘‘Go back to Hong Kong’’
‘‘Travel to see the pyramids’’
‘‘Go on another short-term mission trip to distribute Bibles’’
‘‘Drive across the country’’
‘‘See Auschwitz & Bergen-Belsen’’
Desire to accomplish a personal goal ‘‘Acquire my masters to become a Pathology Assistant’’
‘‘Fly on a private plane’’
‘‘Become a tycoon’’
‘‘Be a backup singer for the Indigo Girls’’
‘‘Get my license’’
‘‘Meet fathers side of the family’’
‘‘I like to be able to swim in spite of age but first I need to know how to swim’’
Downloaded by 191.248.94.42 from online.liebertpub.com at 02/21/18. For personal use only.
4
COMMON ITEMS ON A BUCKET LIST 5
Downloaded by 191.248.94.42 from online.liebertpub.com at 02/21/18. For personal use only.
FIG. 1. Subgroups within patients who have a bucket list. Patients who reported faith, religion, or spirituality to be
important were more likely to have a bucket list compared to those who did not. For those who reported that faith was
unimportant, older (‡61 years) married people were more likely to have a bucket list (v2 = 8.83 and p = 0.07)
have a bucket-list. As having a bucket list is an expression of hope The national attention to precision medicine has recently
and future orientation, this may be the underlying reason. We also been broadened into a focus on precision health37–41 and
found that age influences the bucket list with younger patients personalized care. True personalized care requires that cli-
expressing the desire to complete more daring and risky activities nicians have a clear understanding of what matters most to
compared to older adults. patients and what they wish to accomplish in their lives both
Experts have long recognized the importance of conduct- short term and long term. Asking patients to list their bucket
ing goals-of-care26–32 discussions with patients. However, list is a simple way to elicit their future plans and allows their
current concepts of goals of care remain largely provider- doctors to help them craft a care-plan that will most optimize
centered that is, driven by clinicians’ (and the healthcare the chances of fruition of the desires listed. Knowing the
system’s) needs to have clarity in their patient’s treatment desires listed on their patients’ bucket list will enable clini-
preferences. In a narrow sense, goals of care discussions are cians to go from merely eliciting patient’s preferences about
limited to elicitation of resuscitation preferences. Described specific treatments in a clinical vacuum to anticipating the
more broadly, goals of care discussions constitute a series of impact of such treatments (if any) on the patient’s desired life
conversations about healthcare decisions, specific treatments, goals and plans. As patients progress through their life
the intensity level of care desired and advance care planning. course, their bucket list items will likely change as will their
Conceived this way, goals of care discussions focus exclu- goals of care. Knowing what matters most to patients and
sively on end-of-life treatments choices and fail to capture their bucket lists, clinicians will be able to relate each treat-
what matters most to patients, their desired milestones and ment option to its potential impact on the patient’s life using a
accomplishments in their lifetime, and how they wish to live. personalized approach. This assistance could range from (a) diet
This clinical tendency to approach healthcare-related deci- and exercise counseling for a healthy patient whose life goal is
sions in isolation without connecting them with the patients’ to run a marathon, (b) discussing the strategic timing of an
life goals is likely a key reason why patients may not see the elective knee replacement for an older adult who wishes to
relevance of advance care planning or remain reluctant33–36 dance at his granddaughter’s wedding, to (c) counseling a se-
to complete the related documentation. Many patients, who riously ill patient about treatment benefits (potential life pro-
do not have the health literacy to truly comprehend the im- longation) and burdens (distressing side effects like nausea, hair
pact of their medical decisions on their lives and their family loss) who may then wish to postpone/forego such therapy,
may prematurely choose certain treatment options only to choosing instead to fulfill a bucket list item by traveling to a
change their mind later when they start feeling the real impact final family reunion while still able to do so.
of these choices on their life. We propose the use of the There are limitations to this study. It is a cross-sectional
bucket list to help patients identify what matters most to study of a convenience sample, which limits the generaliz-
them. In contrast to concepts like advance care planning and ability of our findings. As a web-based study conducted in
advance directives, which are not common knowledge to the English, it limits participation of people with limited English
general public, the concept of the bucket list is well known to proficiency, poor technical literacy, and those who do not
many. For example, a Google search on 11/29/2017 yielded have access to a computer. Also, people’s bucket list can and
almost 84 million results in 0.86 seconds for the term ‘‘bucket should change over time, as they get older and also based
list’’ compared to 4.5 million results for the term ‘‘advance on their health status. Tracing changes in the bucket list over
directives’’ in 0.81 seconds and only 533,000 results for the the course of time was beyond the scope of our study. Fur-
term ‘‘advance care planning’’ in 0.57 seconds. thermore, the utility of a bucket list in clinical practice should
6 PERIYAKOIL ET AL.
be studied—for example, will a clinician’s knowledge of the 12. Seidlitz L, Duberstein PR, Cox C, Conwell Y: Attitudes of
patient’s bucket list lead to any changes in the care plan and older people toward suicide and assisted suicide: An anal-
result in more, earlier and richer conversations about the ysis of Gallup Poll findings. J Am Geriatr Soc 1995;43:
patient’s values and life goals? 993–998.
13. https://palliative.stanford.edu (Last accessed September 18,
Conclusion 2017).
Our study shows that diverse patients recognize the con- 14. https://geriatrics.stanford.edu (Last accessed September 18,
cept of a bucket list and are able to articulate a number of 2017).
experiences or achievements that they hope to accomplish 15. http://scopeblog.stanford.edu (Last accessed September 18,
2017).
during their lifetime. Clinicians can elicit the patient’s bucket
16. https://opinionator.blogs.nytimes.com/2015/04/22/we-need-
list and use it as a starting point to initiate goals of care
a-role-reversal-in-the-conversation-on-dying/?mcubz=0 (Last
discussions and as a strategy to craft personalized care plans accessed September 18, 2017).
based on a patient’s own life goals. 17. https://opinionator.blogs.nytimes.com/2015/11/18/pitfalls-
Acknowledgments for-proxies/?mcubz=0 (Last accessed September 18, 2017).
18. www.nytimes.com/2016/09/07/well/family/writing-a-last-
Dr. Periyakoil’s work is funded by R25 MD006857, letter-before-you-get-sick.html?mcubz=0 (Last accessed
NIMHD NIH, U54 MD010724/MD/NIMHD NIH, and September 18, 2017).
AG047366, NIA NIH, Department of Veterans Affairs and 19. www.washingtonpost.com/news/local/wp/2015/04/22/ethnicity-
Stanford University School of Medicine. complicates-patient-doctor-discussion-of-death/?utm_term=.2c2ff
Downloaded by 191.248.94.42 from online.liebertpub.com at 02/21/18. For personal use only.
approach to patient-centred decision-making. Intern Med J 39. Chen Y, Guzauskas GF, Gu C, et al.: Precision health eco-
2014;44:961–966. nomics and outcomes research to support precision medicine:
33. Periyakoil VS, Neri E, Kraemer H: Patient-reported bar- Big data meets patient heterogeneity on the road to value.
riers to high-quality, end-of-life care: A multiethnic, mul- J Pers Med 2016;6: pii: E20.
tilingual, mixed-methods study. J Palliat Med 2016;19: 40. Reischlag JA, Silva MM: Precision health outcomes require
373–379. precise patient identification. JAMA Surg 2016;151:1030–
34. Schickedanz AD, Schillinger D, Landefeld CS, et al.: A 1031.
clinical framework for improving the advance care plan- 41. Cornetta K, Brown CG: Balancing personalized medicine
ning process: Start with patients’ self-identified barriers. J and personalized care. Acad Med 2013;88:309–313.
Am Geriatr Soc 2009;57:31–39.
35. Morrison RS, Zayas LH, Mulvihill M, et al.: Barriers to
completion of health care proxies: An examination of eth-
nic differences. Arch Intern Med 1998;158:2493–2497. Address correspondence to:
36. Knauft E, Nielsen EL, Engelberg RA, et al.: Barriers and Vyjeyanthi S. Periyakoil, MD
facilitators to end-of-life care communication for patients Division of Primary Care and Population Health,
with COPD. Chest 2005;127:2188–2196. Center for Population Health Services
37. Milstein A: Precision health care efficiency via accountable Stanford University School of Medicine
care organizations. JAMA Intern Med 2015;175:1825–1827. 1265 Welch Road
38. Galli SJ: Toward precision medicine and health: Opportu- Stanford, CA 94305
nities and challenges in allergic diseases. J Allergy Clin
Downloaded by 191.248.94.42 from online.liebertpub.com at 02/21/18. For personal use only.